Xenova HF staff commited on
Commit
cf1b882
·
1 Parent(s): cdae1f4

Upload folder using huggingface_hub

Browse files
config.json CHANGED
@@ -3,6 +3,7 @@
3
  "architectures": [
4
  "T5ForConditionalGeneration"
5
  ],
 
6
  "d_ff": 4096,
7
  "d_kv": 64,
8
  "d_model": 1024,
@@ -53,7 +54,7 @@
53
  "prefix": "translate English to Romanian: "
54
  }
55
  },
56
- "transformers_version": "4.29.2",
57
  "use_cache": true,
58
  "vocab_size": 32128
59
  }
 
3
  "architectures": [
4
  "T5ForConditionalGeneration"
5
  ],
6
+ "classifier_dropout": 0.0,
7
  "d_ff": 4096,
8
  "d_kv": 64,
9
  "d_model": 1024,
 
54
  "prefix": "translate English to Romanian: "
55
  }
56
  },
57
+ "transformers_version": "4.34.0.dev0",
58
  "use_cache": true,
59
  "vocab_size": 32128
60
  }
generation_config.json CHANGED
@@ -3,5 +3,5 @@
3
  "decoder_start_token_id": 0,
4
  "eos_token_id": 1,
5
  "pad_token_id": 0,
6
- "transformers_version": "4.29.2"
7
  }
 
3
  "decoder_start_token_id": 0,
4
  "eos_token_id": 1,
5
  "pad_token_id": 0,
6
+ "transformers_version": "4.34.0.dev0"
7
  }
onnx/decoder_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:9ec6e2411cb41f0083b044f58714be7ae91122cbaf9d38401b96512e54fc51ee
3
- size 1874729963
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0aa11d8af73ebdf8ffae811f9e4228d35e3e082138f7b37f1d094dc0819f8dfc
3
+ size 1743133407
onnx/decoder_model_merged.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:b4685637b8ad1ab09ee4a5869bbe33d4d697dec6f17d68c9e7afeaa6c86a177e
3
- size 1875306016
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c9b7c422a23f28cbc365f42597f613e93cf1e411f6ed65f0ab0a30c40f155dd
3
+ size 1743709148
onnx/decoder_model_merged_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:2f5983f2292ee805c533704cf7a01c071955d3405f6b3a475596e7febd38ced1
3
- size 472638601
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6fdc7a0282f60c3ab4a209fb0e2088ce976c22b673eedb8fa625fa706f06b454
3
+ size 439577294
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e9bed6c62e182e0a68c974eac56de272683c812a1ef8785a620b9b124e6e57a6
3
- size 471634408
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4648b0f7301f171fcdb2d4eaa6a48fe9686e7d0244530574fe9a17912040bc43
3
+ size 438573882
onnx/decoder_with_past_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:403f857680db6f799a3983056377193e48912656f3798c59565854883af34453
3
- size 1673339377
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5232a2228e473a538bf8030350774f83df88f5af4b485477acfc9970e573177b
3
+ size 1541742823
onnx/decoder_with_past_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5d9356e0f918a954f6e92cc85d319156dfdfe29a02ba2121d1115a007f8673e0
3
- size 420925549
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:20a12d296d5c99053fac624ba37ddc135fba7589c8204afccc1641bbdaad0c18
3
+ size 387865034
quantize_config.json ADDED
@@ -0,0 +1,139 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "per_channel": true,
3
+ "reduce_range": true,
4
+ "per_model_config": {
5
+ "decoder_model": {
6
+ "op_types": [
7
+ "Softmax",
8
+ "Min",
9
+ "MatMul",
10
+ "Concat",
11
+ "Sqrt",
12
+ "Expand",
13
+ "Transpose",
14
+ "Cast",
15
+ "Tile",
16
+ "Less",
17
+ "Relu",
18
+ "Shape",
19
+ "ReduceMean",
20
+ "Where",
21
+ "Range",
22
+ "Neg",
23
+ "Constant",
24
+ "Sub",
25
+ "Log",
26
+ "LessOrEqual",
27
+ "Mul",
28
+ "Add",
29
+ "Unsqueeze",
30
+ "Div",
31
+ "ConstantOfShape",
32
+ "Reshape",
33
+ "Gather",
34
+ "Pow"
35
+ ],
36
+ "weight_type": "QInt8"
37
+ },
38
+ "encoder_model": {
39
+ "op_types": [
40
+ "Softmax",
41
+ "Min",
42
+ "MatMul",
43
+ "Concat",
44
+ "Sqrt",
45
+ "Transpose",
46
+ "Cast",
47
+ "Less",
48
+ "Relu",
49
+ "Shape",
50
+ "ReduceMean",
51
+ "Abs",
52
+ "Range",
53
+ "Where",
54
+ "Greater",
55
+ "Constant",
56
+ "Sub",
57
+ "Log",
58
+ "Mul",
59
+ "Add",
60
+ "Unsqueeze",
61
+ "Div",
62
+ "ConstantOfShape",
63
+ "Reshape",
64
+ "Gather",
65
+ "Pow"
66
+ ],
67
+ "weight_type": "QInt8"
68
+ },
69
+ "decoder_model_merged": {
70
+ "op_types": [
71
+ "Softmax",
72
+ "Min",
73
+ "MatMul",
74
+ "Concat",
75
+ "Slice",
76
+ "Sqrt",
77
+ "Expand",
78
+ "Transpose",
79
+ "Cast",
80
+ "Tile",
81
+ "Less",
82
+ "Relu",
83
+ "Shape",
84
+ "ReduceMean",
85
+ "Where",
86
+ "Range",
87
+ "Neg",
88
+ "Constant",
89
+ "If",
90
+ "Sub",
91
+ "Log",
92
+ "LessOrEqual",
93
+ "Mul",
94
+ "Add",
95
+ "Unsqueeze",
96
+ "Div",
97
+ "ConstantOfShape",
98
+ "Reshape",
99
+ "Gather",
100
+ "Pow"
101
+ ],
102
+ "weight_type": "QInt8"
103
+ },
104
+ "decoder_with_past_model": {
105
+ "op_types": [
106
+ "Softmax",
107
+ "Min",
108
+ "MatMul",
109
+ "Concat",
110
+ "Slice",
111
+ "Sqrt",
112
+ "Expand",
113
+ "Transpose",
114
+ "Cast",
115
+ "Tile",
116
+ "Less",
117
+ "Relu",
118
+ "Shape",
119
+ "ReduceMean",
120
+ "Where",
121
+ "Range",
122
+ "Neg",
123
+ "Constant",
124
+ "Sub",
125
+ "Log",
126
+ "LessOrEqual",
127
+ "Mul",
128
+ "Add",
129
+ "Unsqueeze",
130
+ "Div",
131
+ "ConstantOfShape",
132
+ "Reshape",
133
+ "Gather",
134
+ "Pow"
135
+ ],
136
+ "weight_type": "QInt8"
137
+ }
138
+ }
139
+ }